Cloning IDS-25s

[wesayso]

For a tweeter solution, without real tweeters, I always wondered what a few of these in between the full rangers could do:

They are called AT (Ambiance Transducer) and come as addition to a full range driver from Audible Physics car speakers and possibly from Melodic Acoustic as well.

They are small enough to put anywhere, wired in line with the full range without any passive components they should provide the air that many full range speakers seem to miss.

Source of pictures: oca123 from DIYMA forums

A description:

Ok I will shed a little light on the workings of the AMT.

First that should not be thought in the sense of a standard tweeter.

AMT in my terms is Air Motion Transducer. Yes yes I know a tweeter is a transducer , but the AMT is more of Ambiance Transducer. For that fact i have be thinking of changing the Name to AAT, Air Ambiance Transducer. it is not an upstage tweeter like the CDT unit.

The way it works, is you simply hook the AMT in parallel with the Nz3-A. No passive crossover or caps need. Why you may ask, well it Because of its natural roll off starting around 19khz, so no cap or crossover are need.

It is simply there to create that what many call that Air sound on the top end that a large cone can not reproduce as well as some dome. But no wide-band can reproduce that Air sound or that last little what is called sparkle as well as a small dome and manufacturer says other wise they are simply selling marketing hype! Now I not saying wide-band can not reproduce highs as there are many that does it extremely well. I sell a few that can do it and no tweeter is need.

So what you get with the Nz3-A/AMT is the best of both worlds the dynamics of the large cone up top and the speed/Air/sparkle of the AMT.

So the to recap they are designed to work together, they are basically one unit just in two pieces. If my degrading memory serve me right, the first of it kind for mobile audio.

Source: Audible Physics NZ3/AMT [Review in progress] - Car Audio | DiyMobileAudio.com | Car Stereo Forum

Never heard them, never tried them, just wondered about mixing in, say, 5 of them in the middle of the array?

 

[njoak]

The IDS-25 clones look superb! Theory as I know it suggests that they would have notably less boominess in the bass due to the way they would interact with room acoustics. I'd be curious to hear your opinion on that.

Thanks. You know, I cross my arrays to separate subs at 80Hz so I only have very limited experience from the bottommost octaves. I did do some very short tests though. With some wall support and moderate listening levels, I'd say they play nicely down to at least 30 Hz (I added a steep HP filter at 30 to be safe). But you can't escape that the headroom is much lower at those frequencies compared to above 150 or so.

Most of all I agree with people that say it is a weird sensation hearing bass that low from such small drivers. And yes, I would say there is an improvement wrt boominess as well. Theory supports it as well - with vertical arrays you don't excite floor-to-ceiling room modes as much. And depending on frequency, speaker placement, and the width of your room there is a fair chance that you can reduce side-to-side room modes as well. Basically only back-to-front modes remain in that case.

 

[njoak]

Njoak,
Beautiful cabinet work there! How are you wiring the 28 drivers as it is not a square number anymore for equivalent series/parallel wiring? You really have to EQ the top end like crazy (+20 dB shelf, Q=0.5, from 15 kHz on up) but that will produce a flat response to 19 kHz. Sounds very good as cone excursion is still very little. You should not need a tweeter array. If you were not using a sub similar EQ on the low end will give bass down to 30 Hz like a 15 inch woofer.

Thanks x! The drivers are 8 ohms nominal, so you can wire them in a 4/7 series/parallell connection to get a total of about 4.7 ohms nominal. That is, sets of 7 drivers in parallell wired to each other in series.

But there is actually a secret feature related to this. The top 3 drivers of each speaker are in a cavity of their own, separated from the remaining 25. Driving them with a separate amp I can use them as Front High speakers that way. The FL/FR channels can be mixed in as well, to make all 28 drivers act as one array without losing the Front High functionality. Pretty clever if you ask me.

I made it easy (well, relatively) to hard-wire the speakers to use only one amp and act as a pure and simple "IDS-28", though. I'll probably do that eventually, since I don't have access to FH decoding in the computer software and kind of lost interest in it anyway.

 

[njoak]

njoak, your arrays are gorgeous!

You got skills!

Now, time to enjoy them!
I concur that arrays are amazing for home cinema.

Thanks perceval! I promise I am enjoying them as often as I can. It seems you will do the same with yours very soon - I am following your thread with great interest!

 

[njoak]

Norman, I have several tracks of guitars ,where it is like the guitar (and player) is in my room in natural size.

Have to agree on this. The sound stage is different and in my setup the impression is often of the musicians being slightly closer to me than the speaker plane, and on a small stage - say half a metre high or so. Difficult to explain, but it is different and larger in a way, without me being able to tell if that is "better or worse". For home cinema I think this works in their favour though, the sound is quite engaging and I feel the imaging is very nice in all 3 dimensions.

I did have the "3 metre _wide_ mouth" syndrome before getting delay and response into some order though, so I think some attention is needed in regards to getting the speakers tuned properly. I will keep tweaking mine...

 

[njoak]

Most of all I agree with people that say it is a weird sensation hearing bass that low from such small drivers.

Correction: that should of course be bass that _deep_, not bass that low...

 

[Bob Richards]

I've read and tend to believe that above about 6kHZ our sense of image location is no longer just left to right (and vice versa) amplitude comparisons. The shape of our ear and audio memory kicks in and also determins image location on the up/down Y axis. I'm not sure, but I suspect that adding one "super tweeter" to a line array (maybe 2/3 the way up) , that only operates above about 7kHZ might integrate in with few negative side effects. Personally I don't mind that the TC and TG drivers roll off at about 15kHZ and have the rolled off response off axis. It does it gradually. But if you want the nth degree of "air", that's what I'd try.

 

[wesayso]

I've been wondering about that but it might be hard to integrate it level wise. The rest of the speaker being a line array and trying to incorporate a point source. Although at high frequencies the full range line doesn't quite add as a line source. So it might work...
Another thought about the use of a super tweeter would be what Lynn Olsen suggests in his Beyond the Ariel thread: up-firing tweeter
I'd recon that would provide the air that some feel missing from a full range line array. But that wouldn't do anything about the distorted vertical image size you get with some recordings. Playing with the measurements of Halair's lines seem to indicate there's something to be had with this idea.

 

[Bob Richards]

I've been wondering about that but it might be hard to integrate it level wise. The rest of the speaker being a line array and trying to incorporate a point source. Although at high frequencies the full range line doesn't quite add as a line source. So it might work...
Another thought about the use of a super tweeter would be what Lynn Olsen suggests in his Beyond the Ariel thread: up-firing tweeter
I'd recon that would provide the air that some feel missing from a full range line array. But that wouldn't do anything about the distorted vertical image size you get with some recordings. Playing with the measurements of Halair's lines seem to indicate there's something to be had with this idea.

Not sure how one might integrate an up-firing tweeter on a floor to ceiling line array, so then it's a matter of how high to mount an additional "super tweeter", and where to cross it over. I also thought it might be better to cross it over at around 15kHZ, where the TG/TC drivers fizzle out, but then the off axis response would have a rolloff from about 3kHZ on up, and then a sudden increase at 15kHZ, which many people would probably think sounds good, but I'm not sure. Too much of the top end can get bothersome after a while. A rear facing supertweeter might work, even if the tower is within inches of the wall, if it's only doing above 7kHZ. The wall might absorb frequencies above 15kHZ a bit, so I'd just keep that in mind.

I guess I figure that I do want at least 15kHZ of bandwidth when I'm sitting in my favorite listening chair position, but off axis I may be less critical and possibly in a conversation with someone, where the highest treble might actually be distracting. I always wondered why most cafes have speakers that have little or no bass, and the highs are rolled off, and I realized that they are less in the way (less distracting) when people are trying to hear each other talk. So there may be something to that.

Another thought; if the "supertweeters" are only doing above 15kHZ, the wavelengths are so short (less than an inch), that you could have two or three supertweeters on the line array, spaced apart enough that the nodal lines created by them would be too fine to be audible (?). I'd put one a third the way down from the top, and another 2/3 the way down, but I would tack them onto the sides of the line array somehow, so I didn't interrupt the consistency of the 3 inch driver line array.

Putting it another way, the cancellations of the comb filter effects would be so close together in frequency, because of the shortness of the wavelengths in comparison with the spacing of the supertweeters, that the frequency resolution of the ear-brain mechanism should not really perceive the cancellations. I could be wrong, but I think this should work. Plus, these tweets are doing such a small amount of bandwidth, they can only do so much damage. Reflections off the ceiling might screw this up a bit.

I might recommend the Fountek 1.5 inch ribbon tweeters for the top end. Their limited off axis dispersion might help them integrate better, and ribbons sound so darn clean at the top end. The waterfall graphs seem to support this. That Zaph guy rated Founteks ribbons as among the best. I've used them with no complaints.

 

[steve f]

After reading everything in this thread, Ive decided to build an IDS-25 clone. What I need are the Speakerbuilder/Audioexpress articles by Roger Russell. Does anyone still have them?

Steve

 

[Olympic]

Steve,

I just built a pair and can provide info. They sound phenomenal!

 

[steve f]

Two members were kind enough to PDF the article.

I've purchased the equalizer, and plan to start assembling the materials in a month.
I'll document my progress and post when I finish or make some real progress. I would enjoy exchanging emails, or post here, with other builders. My very first question is what did you use for grilles? We recently adopted two rescue kittens and I can imagine a big tall speaker becomes a great toy.
Thanks guys.

Steve

 

[bigaudioscotto]

Hey Guys,
I am thinking of building the ids25 using Vifa TC9 drivers,
What Equalizer unit should I look to get to make these work ?

 

[perceval]

Did you check wesayso's build of "Making the Two Towers"?

He talks about equalizers, and also DRC, which really "sounds" like the way to go.

 

[steve f]

Roger Russell, the system designer used a McIntosh MQ107.
John L Murphy designed a similar system and used a Behringer DEQ2496.

I purchased a McIntosh but the Behringer should be much more flexible. It's simply a case of do you want to work analog or digital. I don't know what else is out there perhaps mini DSP?

I haven't started my build yet. The Vifa/Tympany drivers have been used in both paper and woven cone varieties. I'm also impressed by the Dayton Audio ND90-8 and Dayton Audio ND91-8 drivers look good also. I am going to buy a sample of each and measure them. The paper TC9 8 Ohm version and ND90-8 are considerably cheaper considering you need 50 for the design and a couple spares.

I'm hoping that our friends who have built the design will comment.

steve

 

[koldby]

Roger Russell, the system designer used a McIntosh MQ107.
John L Murphy designed a similar system and used a Behringer DEQ2496.

I purchased a McIntosh but the Behringer should be much more flexible. It's simply a case of do you want to work analog or digital. I don't know what else is out there perhaps mini DSP?

I haven't started my build yet. The Vifa/Tympany drivers have been used in both paper and woven cone varieties. I'm also impressed by the Dayton Audio ND90-8 and Dayton Audio ND91-8 drivers look good also. I am going to buy a sample of each and measure them. The paper TC9 8 Ohm version and ND90-8 are considerably cheaper considering you need 50 for the design and a couple spares.

I'm hoping that our friends who have built the design will comment.

steve

As mentioned before, Audiolense is a good solution.
EQ´s the loudspeakers and makes room correction at the same time.

Koldby

 

[wesayso]

It actually depends on how far you want to go. A working solution would be the Behrenger DEQ2496. I have one of those, but never used it. It's a cheap version of more Pro oriented gear. Best would be to go digital in and digital out as the analogue circuit's are their weak link. There will be other pro sound equalisers that are up to this task. Just don't get a graphical equaliser. You need one with lot's of PEQ power available.

I choose to use my PC with JRiver software to use it's EQ and convolution tools. A very powerful tool that corrects more than any stand alone equaliser could ever do. JRiver provides a nice platform. One that can also be made with free tools like foobar with a convolver plugin.
As koldby mentioned, Audiolense is a genuine option to supply the FIR filters, Acourate is another and maybe even Dirac Live could do the job. I use the (awesome) DRC-FIR tools as I am a poor man. But don't think/feel it is less of a solution. Check it out here: DRC: Digital Room Correction

Yet another route that would work is one of the many miniDSP solutions. They even have those with FIR engines, allowing you to use convolution based correction.

So there's lot's of options to choose from. All of them do need a measurement suite (like the free REW) and a (calibrated) measurement microphone.
I use a calibrated Behringer ECM8000 with a IMG Stage Line MPA 102 phantom power/pre amplifier and REW as measurement suite running on a Windows 7 Pro Workstation. But a Dayton Audio UMM-6 USB microphone could also do the job. Lot's more measurement programs available too.

So the amount of control you want dictates the choice. Most people with only a Behringer DEQ unit are as pleased (or maybe even more) as I am. The endless choices as far as processing goes does give you so much to play with, it virtually never ends.

So ask yourself; am I a tinkerer or do I want a setup that's good to go and never touch again . I treat my setup as a playground to learn new things, that steered me to follow the route with the most possibilities.

Hope that helped .

 

[steve f]

I still have a lot to read and learn about DSP. I'm tempted to build Analog first, then after I hopefully get up to speed, go digital.

For an old guy who isn't wealthy, I have to be careful not to make too many false starts. Wesayso, I am the type that never leaves things alone, even when I should. I just love learning new things. The journey is as important as the destination.

In a digital controlled system, I like the idea of not needing the computer connected after set up. I want the system to be as user friendly as possible. That said, besides EQ and a crossover for subs, room correction will probably be a good idea.

I have a lot of homework to do... So guys please share a few links. Thanks.

Steve

 

[wesayso]

Hello Steve,

Here's a write up from forum member Mitchba, showing the needed steps for a correction with Acourate: Computer Audiophile - Acourate Digital Room and Loudspeaker Correction Software Walkthrough

Other systems will work slightly different but the general idea is the same.
Here's a thread with the free DRC, it actually began somewhere on my large build thread:
http://www.diyaudio.com/forums/full-range/272870-drc-trials-failures-successes.html

One of the things I read to get an understanding of the concept was Bob Katz's journey on Audiolense: https://groups.google.com/forum/#!topic/audiolense/f3u_A2aVr70

That should give you some background to get your feet wet.

 

[steve f]

I've had a couple of very serious illnesses in the family and had to put the ids-25 clones on hold. I'm going to try again. I did find that after building a series of test boxes and evaluating several drivers that I preferred both the paper coned and woven coned Vifas to everything else. So did friends and family I bothered by sending them the boxes, then asking a lot of questions.

I have a question for wesayo. One of your posts mentioned that You and Roger corresponded about the differences in the two Vifas. Why did he go with the woven cone? They seem very similar when playing music. I guess I'm wondering if the added cost for woven cones is worth it. Thanks in advance.